切换至 "中华医学电子期刊资源库"

中华损伤与修复杂志(电子版) ›› 2025, Vol. 20 ›› Issue (03) : 199 -205. doi: 10.3877/cma.j.issn.1673-9450.2025.03.003

论著

重度烧伤患者发生早期急性肾损伤危险因素分析及预测模型建立
李培真1,2, 刘海亮1, 李大伟1, 贾昊1, 张泽瑾1, 刘力维1, 申传安1,()   
  1. 1. 100048 北京,解放军总医院第四医学中心烧伤整形医学部
    2. 100853 北京,解放军医学院
  • 收稿日期:2025-03-05 出版日期:2025-06-01
  • 通信作者: 申传安
  • 基金资助:
    国家重点研发计划(2024YFC3016604)

Analysis of risk factors for early acute kidney injury in patients with severe burns and establishment of a prediction model

Peizhen Li1,2, Hailiang Liu1, Dawei Li1, Hao Jia1, Zejin Zhang1, Liwei Liu1, Shen Chuan'an Shen1,()   

  1. 1. Senior Department of Burns and Plastic Surgery,the Fourth Medical Center of PLA General Hospital,Beijing 100048,China
    2. Chinese PLA Medical School,Beijing 100853,China
  • Received:2025-03-05 Published:2025-06-01
  • Corresponding author: Shen Chuan'an Shen
引用本文:

李培真, 刘海亮, 李大伟, 贾昊, 张泽瑾, 刘力维, 申传安. 重度烧伤患者发生早期急性肾损伤危险因素分析及预测模型建立[J/OL]. 中华损伤与修复杂志(电子版), 2025, 20(03): 199-205.

Peizhen Li, Hailiang Liu, Dawei Li, Hao Jia, Zejin Zhang, Liwei Liu, Shen Chuan'an Shen. Analysis of risk factors for early acute kidney injury in patients with severe burns and establishment of a prediction model[J/OL]. Chinese Journal of Injury Repair and Wound Healing(Electronic Edition), 2025, 20(03): 199-205.

目的

分析重度烧伤患者发生早期急性肾损伤(AKI)相关危险因素,并基于危险因素建立风险列线图预测模型。

方法

选取2015年1月至2023年12月解放军总医院第四医学中心烧伤整形医学部收治的烧伤面积≥30%总体表面积(TBSA)的337例患者病历资料进行回顾性分析,并以随机种子1222按照7∶3的比例划分为训练集与验证集。采用最小绝对收缩和选择算子(LASSO)回归法及多因素Logistic回归分析重度烧伤后发生早期AKI的独立危险因素,并构建风险列线图预测模型,以受试者工作特征(ROC)曲线、校准曲线和决策曲线分析(DCA)对模型进行验证。

结果

共121例(35.91%)患者发生早期AKI。体重及Ⅲ度以上烧伤面积大、入院时发生休克、伤后送达医院时间长、入院48 h内血糖和白细胞计数高是重度烧伤患者发生早期AKI的独立危险因素。基于以上6个变量建立重度烧伤患者发生早期AKI风险列线图预测模型,训练集、验证集列线图AUC值分别为0.828(95%CI:0.770~0.886)、0.826(95%CI:0.743~0.909),校准曲线P值分别为0.787、0.125,表明模型预测结果与实际结果吻合较好,DCA表明列线图预测模型具有较高的临床总体净收益。

结论

基于体重、Ⅲ度以上烧伤面积、入院时是否发生休克、伤后送达医院时间、入院48 h内血糖和白细胞计数的列线图预测模型评分,可用于重度烧伤患者发生早期AKI的预测。

Objective

To analyze the risk factors and develop a nomogram-based predictive model for early acute kidney injury (AKI) in patients with severe burns.

Methods

A retrospective analysis was conducted on 337 patients with severe burns (≥30% TBSA) admitted to Senior Department of Burns and Plastic Surgery in the Fourth Medical Center of PLA General Hospital between January 2015 and December 2023.The dataset was randomly split into a training set (70%) and a validation set (30%) using a fixed random seed (1222) to ensure reproducibility.Least absolute shrinkage and selection operator (LASSO) regression and multivariate logistic regression analyses were used to identify predictive variables for constructing the early AKI risk nomogram.The model's discriminative ability, calibration, and clinical utility were evaluated using the receiver operating characteristic (ROC) curve, calibration curve, and decision curve analysis (DCA).

Results

Greater body weight, larger burn area of above third-degree, presence of shock on admission, prolonged time from injury to hospital admission, and higher blood glucose and white blood cell counts within 48 hours of admission were independent risk factors for early AKI in patients with severe burns.The nomogram model, based on these six variables, achieved AUC values of 0.828 (95%CI: 0.770-0.886) in the training set and 0.826 (95%CI: 0.743-0.909) in the validation set.The calibration curve analysis yielded P-values of 0.787 and 0.125, indicating good agreement between predicted and observed outcomes.DCA demonstrated that the nomogram model provided a high net clinical benefit.

Conclusion

The nomogram prediction model score based on body weight, burn area of above third-degree, whether shock occurred at admission, time from injury to hospital admission, and blood glucose and white blood cell counts within 48 hours of admission can be used to predict early AKI in severe burn patients.

表1 AKI组与非AKI组患者基线特征比较
变量 非AKI组(n=216) AKI组(n=121) t/Z/χ 2 >P
体重(kg,xˉ± s ) 70.42±13.48 74.45±12.99 -2.665 0.008
身高(cm,xˉ± s ) 169.46±7.83 170.77±6.89 -1.536 0.125
年龄[岁,M ( Q1,  Q3 )] 39.00( 30.75, 51.00) 44.00( 33.00, 53.00) -2.603 0.009
BMI[ kg/m2M ( Q1,  Q3 )] 24.22( 21.46, 27.04) 25.39( 22.79, 27.68) -2.168 0.030
烧伤总面积[%TBSA,M ( Q1,  Q3 )] 40.00( 35.00, 60.00) 70.00( 40.00, 92.00) -6.460 <0.001
Ⅲ度以上烧伤面积[%TBSA,M ( Q1,  Q3 )] 20.00( 5.00, 30.00) 47.00( 20.00, 75.00) -7.277 <0.001
白蛋白[g/L,M ( Q1,  Q3 )] 33.15( 27.67, 38.12) 28.90( 24.00, 33.60) -4.585 <0.001
血糖[mmol/L,M ( Q1,  Q3 )] 8.25( 6.84, 11.45) 10.93( 8.36, 16.09) -5.303 <0.001
谷丙转氨酶[U/L,M ( Q1,  Q3 )] 28.55( 20.93, 40.15) 38.90( 26.40, 52.00) -4.123 <0.001
总胆红素[μmol/L,M ( Q1,  Q3 )] 18.30( 13.67, 25.12) 24.00( 16.30, 37.70) -4.148 <0.001
直接胆红[μmol/L,M ( Q1,  Q3 )] 6.65( 4.70, 9.60) 8.90( 5.30, 15.00) -3.616 <0.001
血钠[mmol/L,M ( Q1,  Q3 )] 138.00( 136.00, 140.00) 137.00( 134.00, 140.00) -1.480 0.139
血钾[mmol/L,M ( Q1,  Q3 )] 3.92( 3.60, 4.30) 3.92( 3.57, 4.32) -0.649 0.516
白细胞[×109/L,M ( Q1,  Q3 )] 17.02( 13.32, 23.16) 26.76( 17.81, 34.44) -6.321 <0.001
中性粒细胞[×109/L,M ( Q1,  Q3 )] 14.47( 8.97, 20.55) 24.14( 14.96, 30.27) -6.017 <0.001
淋巴细胞[×109/L,M ( Q1,  Q3 )] 1.11( 0.72, 1.80) 1.40( 0.92, 2.15) -2.685 0.007
单核细胞[×109/L,M ( Q1,  Q3 )] 0.89( 0.60, 1.41) 1.41( 0.90, 2.18) -5.132 <0.001
血小板[×109/L,M ( Q1,  Q3 )] 236.00( 184.00, 288.25) 236.00( 179.00, 312.00) -0.766 0.443
NLR[ M ( Q1,  Q3 )] 13.65( 7.01, 22.62) 17.52( 8.51, 26.88) -2.058 0.040
PLR[ M ( Q1,  Q3 )] 207.31( 121.20, 324.34) 163.04( 97.87, 264.91) -2.251 0.024
LPR[ M ( Q1,  Q3 )] 0.48( 0.31, 0.82) 0.61( 0.38, 1.02) -2.260 0.024
MLR[ M ( Q1,  Q3 )] 0.79( 0.55, 1.12) 0.98( 0.64, 1.56) -2.882 0.004
SIRI[ M ( Q1,  Q3 )] 9.27( 5.02, 20.20) 21.41( 7.74, 45.11) -4.344 <0.001
性别[例(%)] 4.991 0.025
155( 71.76) 100( 82.64)
61( 28.24) 21( 17.36)
入院时是否发生休克[例(%)] 46.408 <0.001
197( 91.20) 73( 60.33)
19( 8.80) 48( 39.67)
伤后送达医院时间[例(%)] 18.881 <0.001
0~3 h 164( 75.93) 64( 52.89)
3~6 h 43( 19.91) 46( 38.02)
>6 h 9( 4.17) 11( 9.09)
吸入性损伤[例(%)] 37.477 <0.001
111( 51.39) 34( 28.10)
轻度 62( 28.70) 37( 30.58)
中度 35( 16.20) 22( 18.18)
重度 8( 3.70) 28( 23.14)
是否吸烟[例(%)] 1.272 0.259
147( 68.06) 75( 61.98)
69( 31.94) 46( 38.02)
图1 LASSO回归变量筛选流程。A示28项基线特征的LASSO系数路径图;B示LASSO模型中通过20倍交叉验证选择调谐参数(λ),采用最小化准则进行
表2 重度烧伤患者发生早期AKI多因素Logistic回归分析
图2 重度烧伤患者早期AKI发生概率的列线图 注:列线图使用方法:(1)变量积分计算,针对每个预测变量,从对应数值处绘制垂直线至分数轴,获得该变量的分值;(2)总分转换概率,将各变量的分值相加,从总分处向下绘制垂直线至概率轴,即可得到早期AKI 的发生概率
图3 重度烧伤患者发生早期AKI列线图预测模型的验证。A示训练集ROC曲线;B示验证集ROC曲线;C示训练集列线图校准曲线;D示验证集列线图校准曲线;E示训练集DCA;F示验证集DCA
[1]
Dos Santos RP, Carvalho A, Peres LAB.Incidence and risk factors of acute kidney injury in critically ill patients from a single centre in Brazil: a retrospective cohort analysis[J].Sci Rep,2019, 9(1): 18141.
[2]
Karakaya E, Akdur A, Aydoğan C, et al.A model for acute kidney injury in severe burn patients[ J].Burns, 2022, 48(1):69-77.
[3]
Yue S, Li S, Huang X, et al.Machine learning for the prediction of acute kidney injury in patients with sepsis[ J].J Transl Med,2022, 20(1): 215.
[4]
杨鑫,田新立,刘江,等.预测中重度烧伤患者并发急性肾损伤列线图模型的建立[ J].中华危重病急救医学,2023, 35(7):736-740.
[5]
李晓亮,叶向阳,李琰光,等.肾损伤标志性蛋白在烧伤延迟复苏患者急性肾损伤早期诊断中的价值 [J].中华烧伤杂志,2021, 37(2): 143-149.
[6]
申传安,刘馨竹,李大伟,等.成人大面积烧伤后急救复苏十倍法补液公式的创建与应用 [J].中华烧伤与创面修复杂志,2022, 38(3): 236-241.
[7]
Chen JH, Zhang LW, Liang WJ, et al.The association between systemic inflammatory response index and contrast-associated acute kidney injury in patients undergoing elective percutaneous coronary intervention[ J].Ren Fail, 2024, 46(1): 2330621.
[8]
郭光华,朱峰,黄跃生,等.吸入性损伤临床诊疗全国专家共识(2018版)[ J].感染、炎症、修复,2018, 19(4): 204-209.
[9]
黄跃生.烧伤休克防治全国专家共识(2020版) [J].中华烧伤杂志,2020, 36(9): 786-792.
[10]
Kellum JA, Lameire N.Diagnosis, evaluation, and management of acute kidney injury: a KDIGO summary( Part 1)[ J].Crit Care,2013, 17(1): 204.
[11]
Ostermann M, Bellomo R, Burdmann EA, et al.Controversies in acute kidney injury: conclusions from a Kidney Disease:Improving Global Outcomes( KDIGO) Conference[ J].Kidney Int,2020, 98(2): 294-309.
[12]
Legrand M, Clark AT, Neyra JA, et al.Acute kidney injury in patients with burns[ J].Nat Rev Nephrol, 2024, 20(3): 188-200.
[13]
Bell S, Prowle J.Postoperative AKI-prevention is better than cure?[ J].J Am Soc Nephrol, 2019, 30(1): 4-6.
[14]
孙业祥.中国烧伤休克液体复苏70年进展 [J].中国烧伤创疡杂志,2019, 31(4): 229-235.
[15]
Hsiao KH, Kalanzi J, Watson SB, et al.Adapted approaches to initial fluid management of patients with major burns in resourcelimited settings: a systematic review[ J].Burns Open, 2024, 8(4):100365.
[16]
Bilotta F, Rosa G.Glycemia management in critical care patients[J].World J Diabetes, 2012, 3(7): 130-134.
[17]
Dungan KM, Braithwaite SS, Preiser JC.Stress hyperglycaemia[J].Lancet, 2009, 373(9677): 1798-1807.
[18]
Radzikowska-Büchner E, Łopuszyńska I, Flieger W, et al.An overview of recent developments in the management of burn injuries[ J].Int J Mol Sci, 2023, 24(22): 16357.
[19]
Raup-Konsavage WM, Wang Y, Wang WW, et al.Neutrophil peptidyl arginine deiminase-4 has a pivotal role in ischemia/reperfusion-induced acute kidney injury[ J].Kidney Int, 2018,93(2): 365-374.
[20]
Prasad N, Patel MR.Infection-induced kidney diseases[ J].Front Med( Lausanne), 2018, 5: 327.
[21]
Jiang Z, An X, Li Y, et al.Construction and validation of a risk assessment model for acute kidney injury in patients with acute pancreatitis in the intensive care unit[ J].BMC Nephrol, 2023,24(1): 315.
[22]
Lan J, Xu G, Zhu Y, et al.Association of body mass index and acute kidney injury incidence and outcome: a systematic review and meta-analysis[ J].J Ren Nutr, 2023, 33(3): 397-404.
[23]
Afolabi JM, Kanthakumar P, Williams JD, et al.Post-injury inhibition of endothelin-1 dependent renal vasoregulation mitigates rhabdomyolysis-induced acute kidney injury[ J].Function( Oxf),2023, 4(4): zqad022.
[24]
Bosch X, Poch E, Grau JM.Rhabdomyolysis and acute kidney injury[ J].N Engl J Med, 2009, 361(1): 62-72.
[25]
Su L, Zhang J, Gomez H, et al.Mitochondria ROS and mitophagy in acute kidney injury[ J].Autophagy, 2023, 19(2): 401-414.
[26]
毛自若,周保纯,徐信发,等.早期血乳酸水平预测特重度烧伤患者急性肾损伤加重的价值 [J].中华创伤杂志,2019, 35(6): 556-561.
[27]
Desai RJ, Kazarov CL, Wong A, et al.Kidney damage and stress biomarkers for early identification of drug-induced kidney injury:a systematic review[ J].Drug Saf, 2022, 45(8): 839-852.
[28]
Chen X, Jin H, Wang D, et al.Serum creatinine levels,traditional cardiovascular risk factors and 10-year cardiovascular risk in Chinese patients with hypertension[ J].Front Endocrinol(Lausanne), 2023, 14: 1140093.
[29]
Kjaergaard AD, Teumer A, Witte DR, et al.Obesity and kidney function: a two-sample mendelian randomization study[ J].Clin Chem, 2022, 68(3): 461-472.
[1] 明昊, 宋宏萍, 白誉誉, 党晓智, 赵阳, 程燕妮, 王琪, 肖迎聪. 自动乳腺超声联合临床病理特征预测Luminal B 型乳腺癌术后复发风险的临床研究[J/OL]. 中华医学超声杂志(电子版), 2025, 22(04): 337-347.
[2] 王昕玥, 袁亚, 束华, 曹琨芃, 叶新华, 李璐. 基于LightGBM 算法构建超声引导下凝血酶治疗假性动脉瘤剂量预测模型[J/OL]. 中华医学超声杂志(电子版), 2025, 22(02): 153-161.
[3] 赵晨皓, 张序东, 杨浚沫, 周何. 血清肿瘤标志物对结直肠癌患者术后复发的预测效能研究进展[J/OL]. 中华普外科手术学杂志(电子版), 2025, 19(04): 467-470.
[4] 钱小梅, 罗洪, 李智慧, 周代君, 李东. 76例乙型肝炎肝硬化并发原发性肝癌的高危因素Logistic分析[J/OL]. 中华普外科手术学杂志(电子版), 2025, 19(03): 251-253.
[5] 袁强, 张华宇, 闫璋哲, 朱含放, 陈光, 孙亮, 吕远, 陈纲, 赵锁. cN0峡部偏侧甲状腺乳头状癌中央区淋巴结转移的术前预测模型[J/OL]. 中华普外科手术学杂志(电子版), 2025, 19(03): 333-336.
[6] 朱宗恒, 张志火. 甲状腺乳头状癌对侧中央区淋巴结转移的危险因素分析及预测模型构建[J/OL]. 中华普外科手术学杂志(电子版), 2025, 19(03): 337-340.
[7] 吉林霞, 范小春, 梁琴. 胃癌常见术后并发症类型及危险因素研究进展[J/OL]. 中华普外科手术学杂志(电子版), 2025, 19(02): 220-223.
[8] 杨培容, 潘刚, 周春霞. 胰腺癌术后胰瘘的危险因素及治疗进展[J/OL]. 中华普外科手术学杂志(电子版), 2025, 19(02): 228-230.
[9] 华小玲, 高梦昕, 陈媛, 蔡超, 刘永达, 孙红玲. 良性输尿管狭窄修复重建术研究进展及再手术现状[J/OL]. 中华腔镜泌尿外科杂志(电子版), 2025, 19(03): 377-383.
[10] 张志昱, 张凡, 周奇, 欧阳骏, 林宇鑫, 张学锋. 术前炎性指标对肾透明细胞癌WHO/ISUP病理分级的预测价值[J/OL]. 中华腔镜泌尿外科杂志(电子版), 2025, 19(03): 323-328.
[11] 徐俊洁, 罗虎, 陶锡鹏, 谢李词, 周向东. Ⅰ期非小细胞肺癌经气腔播散的临床、病理及双能量CT 参数与预测模型构建[J/OL]. 中华肺部疾病杂志(电子版), 2025, 18(02): 213-219.
[12] 张宏斌, 杨振宇, 谭凯, 刘冠, 尚磊, 杜锡林. 不可切除肝癌转化治疗后手术的影响因素及预测模型构建[J/OL]. 中华肝脏外科手术学电子杂志, 2025, 14(03): 387-394.
[13] 罗沙, 邱乐乐, 于祥, 冯哲, 蔡广研, 黄静. 无创容量评估新技术在急性肾损伤患者连续性肾脏替代治疗中的应用研究[J/OL]. 中华肾病研究电子杂志, 2025, 14(02): 61-67.
[14] 李菲, 郭晓夏, 郑悦, 郑爔, 李鑫成, 李文雄. 他汀类药物对甘油三酯葡萄糖指数增高的脓毒症相关急性肾损伤患者预后的影响[J/OL]. 中华肾病研究电子杂志, 2025, 14(02): 68-76.
[15] 张佳馨, 刘洋, 吴贞, 程庆砾, 敖强国. 高钠血症对老年院内急性肾损伤患者短期预后的影响[J/OL]. 中华肾病研究电子杂志, 2025, 14(02): 83-90.
阅读次数
全文


摘要